Case Study 2: Fourth Down Decision Analysis for a Championship Team

Introduction

State University finished the regular season 11-1 and earned a spot in the conference championship game. As part of their preparation, the coaching staff wants a comprehensive analysis of their fourth-down decision-making throughout the season.

The analytics department has been tasked with: 1. Auditing all fourth-down decisions for optimality 2. Quantifying points gained/lost from decision quality 3. Identifying systematic tendencies 4. Providing actionable recommendations for the championship game

Background

Team Profile

State University Offense: - Points per game: 34.2 (12th nationally) - Yards per play: 6.4 - Red zone TD rate: 68% - Third-down conversion rate: 44% - Fourth-down conversion rate: 62% (12/15 attempts)

State University Special Teams: - Field goal accuracy: 84% (21/25) - FG accuracy 40+: 75% (6/8) - Punting average (gross): 43.8 yards - Punting average (net): 40.2 yards - Opponent punt return average: 6.8 yards

State University Defense: - Points per game allowed: 18.5 - Yards per play allowed: 4.9 - Red zone TD rate allowed: 52% - Third-down conversion allowed: 32%

Fourth Down Decision Framework

For each fourth-down play, we calculate the expected points (EP) for three options:

  1. Go for it: EP = P(convert) × EP(success) + (1 - P(convert)) × EP(failure)
  2. Field goal: EP = P(make) × 3 + (1 - P(make)) × EP(miss position)
  3. Punt: EP = EP(opponent's starting position after punt)

The optimal decision is the one with the highest expected points.

Season Data Collection

All Fourth-Down Situations (52 total)

Over 12 games, State University faced 52 fourth-down situations:

Category Count
Went for it 15
Attempted FG 25
Punted 12

Detailed Decision Log

Game 1 vs. Eastern State (W 35-14)

Q Time Field Pos Distance Decision Optimal Result
2 8:22 Opp 35 4th & 2 Punt Go Downed at 8
3 4:15 Opp 28 4th & 6 FG (45 yd) FG Made
4 10:02 Opp 42 4th & 1 Go Go Converted

Game 2 vs. Northern Tech (W 28-24)

Q Time Field Pos Distance Decision Optimal Result
1 2:45 Opp 38 4th & 3 Punt Go TB, Opp 25
2 0:45 Opp 25 4th & 4 FG (42 yd) Go Made
3 6:18 Opp 31 4th & 2 Go Go Failed
4 2:12 Opp 18 4th & 3 FG (35 yd) FG Made

Game 3 vs. Midwest University (W 42-17)

Q Time Field Pos Distance Decision Optimal Result
1 5:30 Opp 44 4th & 1 Go Go Converted
2 11:22 Opp 8 4th & G FG (25 yd) Go Made

Game 4 vs. Southern State (W 31-28)

Q Time Field Pos Distance Decision Optimal Result
2 3:45 Opp 36 4th & 4 Punt Go Opp 12
2 0:08 Opp 22 4th & 3 FG (39 yd) Go Made
3 8:15 Opp 45 4th & 2 Go Go Converted
4 1:45 Own 35 4th & 8 Punt Punt Opp 22

Game 5 vs. Western College (W 45-21)

Q Time Field Pos Distance Decision Optimal Result
1 8:02 Opp 32 4th & 1 Go Go Converted
3 5:45 Opp 28 4th & 5 FG (45 yd) Go Made

Game 6 vs. Rival University (L 24-27)

Q Time Field Pos Distance Decision Optimal Result
1 3:22 Opp 41 4th & 3 Punt Go Opp 8
2 5:15 Opp 35 4th & 2 Punt Go Opp 5
3 10:45 Opp 29 4th & 4 FG (46 yd) Go Missed
4 4:22 Opp 38 4th & 5 Punt Go Opp 12
4 0:38 Opp 32 4th & 6 Go Go Failed

Games 7-12 Summary:

Game 4th Downs Went For It Optimal Decisions Key Miss
7 5 1 3/5 Punted 4th & 1 at Opp 45
8 3 1 2/3 FG instead of Go at Opp 33
9 4 2 3/4 Punted 4th & 2 at midfield
10 5 1 2/5 Multiple conservative punts
11 4 2 4/4 None
12 6 2 4/6 FG instead of Go at Opp 30

Analysis

Overall Decision Quality

Season Summary:

Metric Value
Total 4th downs 52
Optimal decisions 32
Suboptimal decisions 20
Decision quality rate 61.5%

Breakdown by Decision Type:

Decision Count Optimal Rate
Went for it 15 13 86.7%
Field goal 25 15 60.0%
Punt 12 4 33.3%

Points Impact Analysis

Expected Points Lost from Suboptimal Decisions:

For each suboptimal decision, we calculate: EP Lost = EP(optimal) - EP(chosen)

Detailed Analysis of Costly Decisions:

Game Situation Chosen Optimal EP Lost
1 Opp 35, 4th & 2 Punt Go 0.8
2 Opp 38, 4th & 3 Punt Go 0.6
2 Opp 25, 4th & 4 FG Go 0.4
3 Opp 8, 4th & G FG Go 0.7
4 Opp 36, 4th & 4 Punt Go 0.5
4 Opp 22, 4th & 3 FG Go 0.5
5 Opp 28, 4th & 5 FG Go 0.3
6 Opp 41, 4th & 3 Punt Go 0.7
6 Opp 35, 4th & 2 Punt Go 0.9
6 Opp 29, 4th & 4 FG Go 0.6
6 Opp 38, 4th & 5 Punt Go 0.5
7 Opp 45, 4th & 1 Punt Go 1.2
8 Opp 33, 4th & 3 FG Go 0.5
9 50, 4th & 2 Punt Go 0.8
10 Opp 40, 4th & 2 Punt Go 0.9
10 Opp 35, 4th & 3 Punt Go 0.8
10 Opp 42, 4th & 1 FG Go 0.6
12 Opp 30, 4th & 2 FG Go 0.7
12 Opp 38, 4th & 4 Punt Go 0.5
12 Opp 44, 4th & 1 FG Go 0.4

Total Expected Points Lost: 12.9 points

Over 12 games, this represents approximately 1.1 expected points lost per game due to suboptimal fourth-down decisions.

Pattern Analysis

Conservative Bias Identification:

Optimal Decision Times Optimal Times Chosen Compliance
Go for it 35 15 42.9%
Field goal 15 25 166.7%*
Punt 2 12 600.0%*

*Values over 100% indicate over-selection relative to optimal

Key Finding: State University shows a strong conservative bias: - Goes for it only 43% of the time when optimal - Kicks field goals 67% more often than optimal - Punts 6x more often than optimal

Situational Tendencies

By Field Position:

Zone 4th Downs Go Rate Optimal Go Rate Gap
Opp 0-10 4 50% 75% -25%
Opp 11-25 12 17% 50% -33%
Opp 26-40 18 28% 78% -50%
Opp 41-50 10 40% 90% -50%
Own 40-50 5 40% 60% -20%
Own 0-39 3 0% 0% 0%

Greatest Gap: The opponent's 26-50 zone shows the largest discrepancy between actual and optimal go-for-it rates. This is the "no-man's land" where the team is too far for a comfortable field goal but too close to give up on scoring.

By Distance:

Distance 4th Downs Go Rate Optimal Go Rate Gap
4th & 1 12 58% 92% -34%
4th & 2 14 36% 79% -43%
4th & 3 10 20% 60% -40%
4th & 4 8 25% 50% -25%
4th & 5+ 8 13% 25% -12%

Key Finding: Short-yardage situations (4th & 1-2) show the largest missed opportunities. The team should go for it ~85% of the time on 4th & 1-2 in opponent territory.

By Score Differential:

Score Context 4th Downs Go Rate Notes
Leading 14+ 10 10% Very conservative when ahead
Leading 1-13 15 27% Moderately conservative
Tied 12 33% Closer to optimal
Trailing 1-13 11 45% More aggressive when behind
Trailing 14+ 4 75% Forced aggression

Pattern: The coaching staff becomes appropriately aggressive only when trailing significantly. Conservative play while leading may be costing opportunities to extend leads.

Game 6 Deep Dive: The Loss

The only loss came against Rival University, 24-27. Let's examine the fourth-down decisions:

Situation 1: Q1, 3:22, Opp 41, 4th & 3 - Decision: Punt (downed at Opp 8) - Optimal: Go for it - EP Analysis: - Go: 0.56 × 2.3 + 0.44 × 1.2 = 1.82 - Punt: 0.9 (opponent at their 8 = -0.9 EP for them = +0.9 for us) - EP Lost: 0.92 - Result: Opponent drove 92 yards for touchdown

Situation 2: Q2, 5:15, Opp 35, 4th & 2 - Decision: Punt (downed at Opp 5) - Optimal: Go for it - EP Analysis: - Go: 0.62 × 2.0 + 0.38 × 0.8 = 1.54 - Punt: 0.6 - EP Lost: 0.94 - Result: Opponent punted, field position battle

Situation 3: Q3, 10:45, Opp 29, 4th & 4 - Decision: FG (46 yards, missed) - Optimal: Go for it - EP Analysis: - Go: 0.48 × 2.5 + 0.52 × 0.5 = 1.46 - FG: 0.65 × 3 + 0.35 × 1.2 = 2.37 - Actually, with 65% make rate, FG might be optimal here...

Let me recalculate. The kicker had shown 75% on 40+ yard kicks (6/8). Using team-specific rate: - FG: 0.75 × 3 + 0.25 × 1.2 = 2.55

This was actually close to optimal. However, the miss was costly.

Situation 4: Q4, 4:22, Opp 38, 4th & 5 - Decision: Punt (downed at Opp 12) - Optimal: Go for it - EP Analysis: - Go: 0.42 × 2.1 + 0.58 × 0.9 = 1.40 - Punt: 0.7 - EP Lost: 0.70 - Context: Down 3 points with 4+ minutes left - aggressive play warranted

Situation 5: Q4, 0:38, Opp 32, 4th & 6 - Decision: Go for it (failed) - Optimal: Go for it - Analysis: Correct decision, unsuccessful execution

Game 6 Summary: - Total EP lost from suboptimal decisions: ~2.5 points - Final margin: 3 points - Conclusion: Conservative decision-making may have directly contributed to the loss

Championship Game Recommendations

Based on season-long analysis, here are specific recommendations:

Recommendation 1: Increase Aggression in No-Man's Land

Current: 28% go-for-it rate at opponent's 26-40 Recommended: 65%+ go-for-it rate at opponent's 26-40

Specific Rules: - 4th & 1-2 at opponent's 26-40: Go for it unless trailing by 7+ in final 2 minutes - 4th & 3 at opponent's 26-40: Go for it if not in FG range (<45 yards) - 4th & 4 at opponent's 26-40: Evaluate based on down/distance conversion history

Recommendation 2: Short Yardage Aggression

Current: 58% on 4th & 1, 36% on 4th & 2 Recommended: 90%+ on 4th & 1, 75%+ on 4th & 2

Justification: - Team's conversion rate: 62% overall, higher on short yardage - Expected value strongly favors going for it in most field positions - Risk of punt/FG leaving points on the field

Recommendation 3: Reduce Punting from Opponent Territory

Current: 12 punts from opponent territory all season Recommended: Maximum 2-3 punts from opponent territory per season

The only justified punts from opponent territory: - 4th & 10+ with no FG opportunity - Leading in final 3 minutes and pinning deep is strategically valuable - Weather conditions severely limiting conversion probability

Recommendation 4: Context-Specific FG Decisions

When to kick field goals: - Inside opponent's 25-yard line when distance is 4th & 5+ - When FG wins or ties the game in final minutes - When weather significantly reduces conversion probability

When to go for it instead of FG: - 4th & goal from inside the 5-yard line - 4th & 1-3 from opponent's 25-35 (outside comfortable FG range) - When leading by 4-7 points (TD extends lead more than FG)

Recommendation 5: Championship Game Cheat Sheet

Prepare a printed reference for the sideline:

Field Position 4th & 1 4th & 2 4th & 3 4th & 4 4th & 5+
Opp 1-5 GO GO GO GO FG
Opp 6-15 GO GO FG FG FG
Opp 16-25 GO GO GO FG FG
Opp 26-35 GO GO GO GO PUNT
Opp 36-45 GO GO GO GO PUNT
Opp 46-50 GO GO GO PUNT PUNT
Own 40-45 GO PUNT PUNT PUNT PUNT
Own <40 PUNT PUNT PUNT PUNT PUNT

Adjust for score differential and time remaining

Expected Impact

If State University implements these recommendations:

Conservative Estimate: - Optimal decision compliance: 61.5% → 80% - EP saved per game: 0.5 points

Aggressive Estimate: - Optimal decision compliance: 61.5% → 90% - EP saved per game: 0.9 points

Championship Game Impact: - Against a quality opponent, 0.5-0.9 EP could be the difference - Over a 3-game playoff run, 1.5-2.7 total EP gained - Translates to approximately 10-15% improvement in close-game win probability

Conclusion

State University's fourth-down decision-making shows a clear conservative bias that cost approximately 12.9 expected points over the season. The lone loss of the season came in a game where conservative fourth-down decisions cost an estimated 2.5 points in a 3-point defeat.

For the championship game, implementing a more analytically-driven approach to fourth downs could provide a meaningful competitive advantage. The team should:

  1. Go for it more often in opponent territory
  2. Be especially aggressive on short yardage
  3. Trust their high conversion rate (62%)
  4. Use a decision reference chart to remove in-game hesitation

The combination of a strong offense (62% conversion rate) and strong defense (limits damage from failed conversions) makes aggressive fourth-down play even more advantageous for State University than it would be for an average team.

Code Implementation

See code/case-study-code.py for: - FourthDownAuditor class - Expected points calculations - Decision optimization analysis - Visualization generation

Discussion Questions

  1. How should the coaching staff balance analytical recommendations with their "feel" for the game situation?

  2. The analysis assumes State University's conversion rate (62%) will hold in the championship game. How would you adjust recommendations if facing an elite defense?

  3. Should fourth-down aggressiveness change based on the opponent's offensive quality? (A failed attempt against a potent offense has higher expected cost)

  4. How would you present this analysis to a coaching staff that may be skeptical of aggressive fourth-down strategies?

  5. What additional data would improve the analysis? (Time remaining impact, specific defensive personnel, weather forecast, etc.)